Skin, Kin and Clan
This chapter looks at the evidence for the close–distant dichotomy in the kinship systems of Australian Aboriginal societies. The close–distant dichotomy operates on two levels. It is the distinction familiar to Westerners from their own culture between close and distant relatives: those we have frequent contact with as opposed to those we know about but rarely, or never, see. In Aboriginal societies, there is a further distinction: those with whom we share our quotidian existence, and those who live at some physical distance, with whom we feel a social and cultural commonality, but also a decided sense of difference. This chapter gathers a substantial body of evidence to indicate that distance, both physical and genealogical, is a conception intrinsic to the Indigenous understanding of the function and purpose of kinship systems. Having done so, it explores the implications of the close–distant dichotomy for the understanding of pre-European Aboriginal societies in general—in other words: if the dichotomy is a key factor in how Indigenes structure their society, what does it say about the limits and integrity of the societies that employ that kinship system?
Kinship is synonymous with anthropology. Morgan’s (1871) Systems of Consanguinity and Affinity of the Human Family is one of the founding documents of the discipline. It also has an immediate connection to Australia: one of the first fieldworkers to assist Morgan in gathering his data was Lorimer Fison, who, later joined by A. W. Howitt, began the task of investigating the kinship systems of Australian Indigenes. Since then, Australian kinship has often been at the forefront of anthropology’s theoretical investigation of its meaning and significance, most notably in the work of Lévi-Strauss (1969). Australian kinship has provided case material for expositions on many innovative approaches to the study of kinship structure and logic, such as Scheffler’s (1978) application of extensionist theory to Australian systems. In short, in the study of Australian kinship, there can be traced a veritable history of kinship study taken as a whole, either as evidence in the advancement of new theories or by the application of theory developed from societies elsewhere in the world to Australian societies (e.g. Meggitt’s 1962 use of ‘descent theory’ in respect to the Warlpiri; see Kuper 2005). All of these approaches can be said to share a common characteristic: they are theories developed by anthropologists who have then applied them to their Australian subject matter. This chapter seeks a different approach: it will examine the evidence for a particular theoretical model that appears not to have had its origins in anthropological theorisation, but is emic, intrinsically Indigenous, the presence of which in ethnography can be attributed solely to having been observed in, or elicited from, Aboriginal informants.
I am aware of only one study of Australian kinship that has recognised the centrality of the close–distant dichotomy in Aboriginal kinship: D. H. Turner’s (1980) Australian Aboriginal Social Organization. As will be demonstrated, the close–distant dichotomy in the Indigenous conceptualisation of kinship was observed from the outset of the study of Australian societies, and its recurrence in ethnographic description since has been a persistent motif. This chapter seeks to examine explanations for this ubiquity. The first task is to determine whether the dichotomy is a genuine Indigenous conception. The chapter is to discover if the dichotomy is universal in Aboriginal kinship, if it has the same degree of significance in all systems, and if the role it plays in kinship systems is identical or varies from system to system.
If, as is contended, the close–distant dichotomy is an Indigenous conception, then we need to find the value of this for interpretation: should we assume the dichotomy is interesting, but ultimately irrelevant to the understanding of kinship systems? Or do we conclude that the Indigenous view ought to be privileged, not only because it is the Indigenous view, but also because it provides otherwise unobtainable insight into the nature of the kinship-structured society? My sympathies lie with the latter proposition, as expressed by Needham (1962, p. 259) in his comments on the difficulties he and his colleagues experienced in trying to understand the Wikmunkan system of Cape York:
The source of all our analytical difficulties may be traced to a single factor, namely, a failure to apprehend Wikmunkan social life through the categories and connecting ideas of the peoples themselves. Instead, their social organization was conceived and described in terms of the concern for genealogical connection which is habitual to a European observer but which is fundamentally mistaken in understanding a society of this kind.
While understanding the Indigenous reasoning behind kinship structure may provide a corrective for the mistaken notion that a society’s kinship can alone be understood by the application of Western analytical concepts, the social implications of Indigenous reckoning still remain. To suppose that the close–distant dichotomy was employed by its practitioners with a consciousness of its wider anthropological implications, questions regarding the physical and social limits that kinship structure imposes, the integrity or homogeneity that a kinship system implies for a society, historical development, and so on, are unlikely. If, as surmised, the close–distant kin dichotomy was a ubiquitous feature of Aboriginal societies across their distribution, then an institution of this significance ought to reflect these broader implications. The chapter concludes with a brief exploration of the questions that this dichotomy raises for a wider understanding of Aboriginal social organisation.
The Close–Distant Dichotomy: A Short History
The first, most central question, whether or not the dichotomy is a genuinely Indigenous idea, is not easy to answer. Some anthropologists have claimed outright that it has no role in the systems they have studies. At least one authority, Turner (1980, p. viii), claimed universality for the principle. In any event, its recurrence in ethnographies from the discipline’s beginnings, even in those that programmatically adhere to conventional forms of analysis, requires explanation. In most cases, the role of the close–distant dichotomy falls into somewhat of an ambivalent category. It is spelled out in very few ethnographies as a means through which people themselves understand and construct their relationships. And yet, it is not, like descent or affinal theory, a well-discussed and commented upon part of the anthropological lexicon of kinship. It is fair to say that the dichotomy usually appears by default, as a ubiquitous recourse for explanation, often with the explicit or inferred understanding that it has been derived from informants. It has rarely been discussed as a principle as such.1 Textually, this ambivalence expresses itself in the frequent use of inverted commas around the words ‘close’ and ‘distant’ (see Turner 1980, p. viii). Again, it is difficult to know whether this is because the anthropologist has heard it expressed in these words from their informant, or whether it is an acknowledgement of its unorthodoxy in scholastic usage. However, it is clear that from the outset of professional anthropology, it is a distinction that has been observed and recorded in the field. Radcliffe-Brown (1930, pp. 2, 236), for example, described the process of betrothal in the Kumbaingeri system as follows:
Marriage is prohibited with one’s own mother’s brother’s daughter, or father’s sister’s daughter. A man marries a woman who belongs to the same section and generation as his mother’s brother’s daughter, and who is, according to the terminology, a relative of the same kind. But she must come from another part of the country, and must not be closely related to him. The normal procedure was described to me as follows. A woman who is ‘father’s sister’ to a boy, possibly his own father’s sister, would look out for a wife for him. Finding a woman who was her ‘sister’, but not closely related to herself or her nephew, she would induce the latter to promise her daughter in marriage to the boy. From this moment this woman becomes the boy’s mother-in-law, and he must avoid her. It is, therefore, preferable that he should never have met her before the arrangement is made [empahsis original].
Elkin (1937–38), like Radcliffe-Brown, made rare, yet specific, references to the close–distant dichotomy. I have found only one occasion in which Elkin generalised the ubiquity and force of the distinction in Aboriginal kinship:
The tendency amongst the Australian Aborigines is to select the mother-in-law rather than the wife, and to seek her as far away as possible both in geographical position and relationship, on account of the avoidance associated with her and her parents. This is even noticeable among many tribes in which second-cousin marriage is permissible, with the result that this marriage tends to be rare … by seeking a more distant wife’s mother than own father’s sister, father’s mother is not drawn into the relationship of wife’s mother’s mother.2 (Elkin 1937–38, p. 432)
In several instances, Elkin provided good evidence that the dichotomy was acquired firsthand from his informants. For example, in quizzing his Arabana informants on the workings of their system, Elkin (1937–38, p. 441) encountered the following response: ‘When first speaking to them I received the impression that a man could marry his cross-cousin, but when they realized that the woman concerned was the daughter of one’s own mother’s brother or own father’s sister, they protested in decided terms that such a marriage was impossible’. The reason for this misunderstanding, in the informant’s terms, is later made clear:
With regard to mother’s mother’s brother’s daughter’s daughter, my informants stated that she might be either bilya, the term which is also applied to mother’s brother’s daughter and with whom marriage is prohibited, or nupa, wife; but when they realized that I meant own second cousin, some hesitated and even denied the possibility of such marriage, saying that such a woman was too close, ‘all one relation’ and that she came from a man’s own kadini, mother’s mother’s brother. (Elkin 1937–38, p. 443)
From this, Elkin (1937–38, pp. 442–3) drew the following conclusion in respect to Arabana affinal relationships: ‘A man may marry a woman called nupa, the daughter of a “distant” kagaga and ngauwili, mother’s brother and father’s sister, who are distinguished from own mother’s brother and father’s sister by being called kagaga taru and ngauwili yambua’.
Elkin (1938–39, p. 45) provided several examples throughout his Oceania series ‘Kinship in South Australia’ of the way in which the close–distant dichotomy melded in with other aspects of kinship and social organisation. For example, the Wilyakali have ‘special terms’ to distinguish ‘own’ from distant kin. There are several instances of how the distance distinction overrides or negates totemic relationships: ‘As far as kinship rules went, this marriage was quite in order, but a far distant social totemic relationship was found to exist between them, “which make them brother and sister”. The kindred, however, decided that, as this relationship was a far distant one, it could be ignored’ (Elkin 1938–39, p. 52). Additionally, there are case studies, particularly of Western Desert (Aluridja) informants, that highlight the application of the close–distant dichotomy in the calculation of relationships with individuals, otherwise strangers, upon their entering the ‘close’ community. The possibility that Western Desert culture provides a somewhat distorted perception of the importance of the close–distant dichotomy to Australian kinship, by virtue of the special physical conditions that apply, will be returned to later (see Elkin 1932, pp. 304–5 quoted in Turner 1980, p. ix, in respect to the Karadjeri).
In a very different environment, Stanner (1936–37) examined the kinship of the Murinbata of the coastal tropics, specifically the evolution of their system as it adjusted to the eight-subsection system introduced from the Djamindjung to their south. Like Elkin, Stanner (1936–37, pp. 197–8) had recourse to distance in the determination of kin terms, particularly those with direct application to marriage: ‘Marriage with own cross-cousins was prohibited. These marriages were effectively prevented by extending the terms for mother (kale) and mother’s brother (kaka) to one’s own cross-cousins, but not to the children of more distant mother’s brothers and father’s sisters. One’s own cross-cousins were “little kale” and “little kaka”’. This distance-based distinction was still effective some 40 years later when Falkenberg and Falkenberg (1981, p. 175) resumed Stanner’s work on Murinbata kinship: ‘A man should not marry a pugali who is the daughter of kaka ngoitnan in his mother’s local clan, but only a pugali who is the daughter of kaka ngoitnan “from far away”, i.e. from another local clan. Further, a man should not marry a pugali who is the daughter of bip:i ngoitnan from his own local clan, but only a pugali who is the daughter of bip:i ngoitnan “from far away”’; and ‘A Murinbata does not distinguish terminologically between those pugali whom he cannot marry and those who are eligible as his wives, but when such a distinction is desirable or necessary he will refer to the former as pugali and to the latter as pugali pugali, i.e. pugali “from far away”, who are the children of kaka ngoitnan and bip:i ngoitnan from alien clans’ (Falkenberg & Falkenberg 1981, p. 178).
Similarly, Hiatt (1965, p. 78), however working with the Gidjingali of the Blythe River of Arnhem Land, another well-endowed tropical environment, acknowledged the role of the dichotomy, without arriving at any understanding of its basis:
The only men with rights to their wives were those married to the six women in class A. I have distinguished class B from the others because giving a man his FZD was a recognised subsidiary to orthodox bestowal. In distinguishing close from distant relationships I have assigned marriages to classes C or D if, as well as having the appropriate classificatory relationship, the partners are linked as cognates (MFZDD, MBD, &c.). This corresponds roughly with a distinction made by the natives themselves, who spoke of ‘close’ and ‘distant’ connections but did not apply any strict criterion.
Generally speaking, the relationship between genealogical distance and spatial distance is ambiguous. Radcliffe-Brown (1930, pp. 438–9) commented: ‘When natives speak of “distant” relatives they combine in the one conception both genealogical remoteness and geographical distance’. For the anthropologist, therefore, getting the balance right in the implication of distance is not always easy. Turner (1980, p. viii) drew attention to the fact that expressions of kinship distance in English—‘close’, ‘far-away’, ‘near’, ‘distant’ and so on—are ‘strictly genealogical’ and ‘may have unfortunate implications for our understanding if we translate them directly’. One could argue the term ‘classificatory’ comes into use as a corollary for distance—that is, the determination of distant kin relationships is founded on an abstract (and sometimes variable) structure of quasi-genealogical relationships (McConnel 1933–4, p. 350; Sutton 1978, p. 199). Conversely, ‘close’ kin have a greater claim to a biological relationship (although this too, as the evidence suggests, is hardly a fixed principle), while with kinfolk who are ‘distant’, the possibility of a genealogical relationship based on biology diminishes, and relationships are not genealogical but ‘classificatory’.3 This is a distinction that in one form or another goes back to the earliest ethnographic work in Australia. Howitt (1904, p. 161), for example, contrasted one’s ‘own’ children from one’s ‘tribal’ children, where ‘tribal’ is a synonym for ‘classificatory’ (see Radcliffe-Brown 1913, p. 158). In Stanner (1936–37, p. 199), we see the ready potential for substitution of the terms ‘distant’ and ‘classificatory’:
The term pugali has been taken over from the Djamindjung to denote those cross-cousins who could not have been married under the former Murinbata system, and may still not be married under the altered system. Distant cross-cousins referred to as ‘half pugali’ whose subsections are appropriate, may be married, but they are then called by the normal Murinbata term for wife (purima). This was the term formerly applied only to classificatory mother’s brother’s daughters and father’s sister’s daughters … It is worth noting that the children of female pugali are being called wakal nginar, a term formed by the suffixation of nginar (mother-in-law) to the ordinary bisexual term for child (wakal). The wakal nginar is distinguished from the pipi nginar, who is the classificatory father’s sister (pipi). Both these women give their daughters to a man.4
Similar to the correlation of distant kin with classificatory kin is the equation by both Stanner and Falkenberg of distant kin with ‘different hordes’, ‘alien local clans’ and similar expressions, as well as specific references to entities such as the ‘mother’s clan’. It is a reasonable inference that no matter how attenuated the role of physical distance becomes in the determination of kinship relationships beyond one’s patrigroup, the close–distant principle remains equally effective. Turner (1980, p. ix) quoted R. M. and C. H. Berndt (1970, p. 87) on the Gunwinggu, for whom ‘even though genealogical proximity is significant’ in reckoning ‘closeness’, ‘it can be offset by other factors. One is territorial affiliation. Two men from the same or adjacent small territories or cluster of named sites are “brothers” even if no genealogical links can be traced. Each is “close father” to the other’s sons, and may be acknowledged as “closest father” if no “father” from a common grandparent is living … The fact of being neighbours is important in itself, but mythical and ritual connections are even more so’. The parameter of distance in these more closely knit coastal societies is not absolute; there is no determinative relation between distance and the desirability of alliance, and, in fact, the opposite tendency may apply—the desire for alliance with immediate neighbours. In these more densely populated areas, the desire for broad interrelationships is balanced by the desire to preserve the strength of a core group that is both genealogically and spatially proximate.
McConnel’s (1933–34) work on the Wik-Mungkan of Cape York drew very similar conclusions to that of Stanner, as we perhaps might expect from people of two reasonably similar environments. Like Stanner (1935–36, 1936–37) and Falkenberg and Falkenberg (1981), the use of the terms ‘close’ and ‘distant’ appears throughout:
It often happens that muka [FeB] and kala [MyB] in the mother’s clan may be married to pinya [FyZ] from more distant clans than the father’s; also, that pinya in the father’s clan may be married to muka and kala in more distant clans than the mother’s. A distant pinya is not compelled to give her daughter to her husband’s sister’s son—nor is a muka from a distant clan entitled to insist upon the marriage of his son with his wife’s brother’s daughter. The element of choice enters into the situation. (McConnel 1933–34, p. 341)
The ‘element of choice’ McConnel referred to is the scale of desirability in the contract of affinal relationships already commented on in Stanner’s and Falkenberg’s analyses of the Murinbata, and by the Berndts in respect to the Gunwinggu. While spatially distant relationships are sought after for certain reasons, so too is the maintenance or extension of relationships with clans with whom connection already exists. McConnel (1933–34, p. 341) concluded that:
A pinya from a distant clan may wish her daughter to marry back into ‘company’ clans in her own locality with which her clan has older connections and more urgent obligations. She may, however, prefer her daughter to marry her husband’s sister’s son, since she must live in her husband’s locality, and would like to keep her daughter near her. In this case she will ‘promise’ her daughter to her sister’s son.
As a consequence, a local group will be composed of women who have married in whose ‘common kattha [M] may hail from a number of different clans near and far’ (McConnel 1933–34, pp. 330–1). This results in the local clan being able to exercise a number of options in the alliances it wishes to contract with other clans—both ‘close’ and ‘distant’ (which are, of course, relative determinations) (McConnel 1939–40, pp. 448–9). Similarly, Sutton (1978, p. 106) observed in the coastal Wik groups that marital partners were usually preferred with kin on the ‘closer’ end of the physical distance scale:
The preferred marriage is that between classificatory (non-actual) cross-cousins of the type MBD = FZeS whose clan estates are in close proximity. I will later show that there is a tendency for marriages to form regional clusters which may be defined by two major parameters, the inland/coastal distinction and ritual group membership.
This ‘preference for marriage with near neighbours’ was the result of a strong political tendency among the coastal Wik:
Local endogamy is politically motivated. It binds the local groups of an area into mutual support when threatened from outside on any scale, and reduces conflict at the local level. This is a conscious policy and stated quite clearly and often by my informants … The social bond between affines of a locality make for military strength. It is often mentioned in descriptions of fighting that one group were munhtha-mooerinhthjanha, a compound term denoting a set of cross-cousins. (Sutton 1978, p. 130)
Among the Wik, the desire to acquire kin at a distance was correspondingly weaker. However, as with the inland Wik-Mungkan, alliances with distant kin could still be contracted if a political advantage or ambition was served (Sutton 1978, pp. 83–4), and, as with inlanders, marriage contracted at distance appeared more often to involve the direct exchange of ‘sisters’ (see McConnel 1939–40, pp. 451–3). With coastal Wik, ‘for demographic reasons those kin who are more distant genealogically tend also to be those kin who come from more distant places and from groups who are politically disjunct from those of ego’ (Sutton 1978, p. 199).
The role played by physical distance can be seen to vary considerably between coastal peoples (or, at least, it might be assumed, among peoples with relatively high levels of population density) and those where this is less the case. In Australian Aboriginal Social Organization, Turner (1980, p. 7) identified a contrasting tendency between societies whose application of the distance dichotomy was to achieve close or progressively more distant relationships ‘between the benefits of patri-group endogamy or of ever-expanding “patri-group family” exogamy and their associated “brotherhood” types’, and sought to raise this distinction to a general principle. It is a dichotomy he explored by comparison between, at the endogamous extreme, the Kaiadilt of Bentinck Island and the Warnungmanggala of Groote Eylandt, both in the Gulf of Carpentaria, and at the other, exogamous pole, the Yaralde of coastal South Australia. While ‘exogamy outside one’s own and male ancestors’ “patri-group families … would not create such close ties within the patri-group as endogamy [it] would achieve the widest possible range of organic-like relations and most comprehensive network of mechanical ties with the larger society”’ (Turner 1980, p. xi). Conversely, less expansive alliances ‘are thereby able to achieve a degree of solidarity between a number of patri-groups through intermarriage within a relatively small circle … The security achieved, though not covering as many contingencies as under the Yaralde system nor allowing for as intimate a knowledge and as efficient an exploitation of the local area as, theoretically, under endogamy, nevertheless would be considerable within the collective estates of a small number of groups whose members engaged in constant intercourse and exchange and thus, as a collective, formed a geographically continuous population’ (Turner 1980, p. 7).
The Western Desert
The coastal Wik, and other densely populated societies, perhaps represent one extreme of a scale of distance reckoning that extends at the other end to the Western Desert peoples, for whom alliance at physical distance was a premium. No doubt this contrast was predicated by the different environmental and political conditions that animated these societies. As Western Desert life was physically uncertain and critically variable, connection over distance was imperative (see Smith 2013, pp. 296–98, 329–30). However, this was more than simply a matter of survival, as the historically rapid expansion of the Western Desert people across their vast distribution must surely have been facilitated by their ability to readily (if not periodically) coalesce and operate as a collective when necessary (see Elkin 1939–40, p. 203; McConvell 1996; Myers 1986, pp. 155–6, 159). For coastal dwellers such as the Wik, where one’s own and adjacent countries provided the greater part of the necessities of life, relationships at distance were far less pressing. Diametrically opposite political necessities also applied; the extension of influence might be desirable in the desert, but defence of one’s own well-endowed country through strong local connections was the overriding concern for coastal peoples.
For the unique circumstances of their environment, Western Desert kinship provides a forum for some of the chief issues that arise from the close–distant dichotomy: its Indigenous origin and the relationship between genealogical and physical distance, and the social context of the dichotomy. From the outset, ethnographers have recorded the close–distant dichotomy in Western Desert society. Sackett (1976, p. 139) recapitulated the discipline’s state of knowledge up until his fieldwork:
Aborigines throughout the Western Desert state the prescribed rule of marriage is to a cross-cousin ‘a little bit far away’ (Berndt & Berndt 1945, p. 151; 1964, pp. 70–4; Fry 1950, p. 290; Yengoyan 1970a, p. 85), making a system which has been termed a variant of the Kariera practice. As Radcliffe-Brown (1931, p. 439) and Piddington (1970, p. 342) note, the distance implicit in this rule is conceptualized in genealogical as well as spatial terms. In other words, ego must not marry a relative from his own local group or an actual cross-cousin from another country. In all likelihood, a spatially close cross-cousin would also be an actual MBD/FZD, though the converse need not be true.
Numerous sources have since made clear that the close–distant dichotomy is the guiding principle in the determination of kin relationships for the Western Desert people, one that has emerged from within the culture itself. Myers (1986, p. 175) recorded the following:
Distance is the key, as one young Papunya man made clear in explaining why he could not marry a girl he admired from his own settlement. They were, he said, ‘from one ngurra’. In the Pintupi view, they were ‘too close’ (ngamutja ‘from nearby’), and one’s spouse must be ‘from far away’ (tiwatja).
Although not kin terms, this specific terminology is used to differentiate close and distant kin takes us one step closer to the idea that the distance is at the centre of Western Desert kinship structure. Myers (1986, p. 195) stated outright that kin categories were also based on a physical interpretation of the close–distant dichotomy:
In deciding how to classify individuals in kin terms, Pintupi regularly make a distinction between ‘close’ and ‘distant’ kin that has an important impact on classification. This distinction effectively makes locality another criterion of the kinship system. The isomorphic relationship between being ‘close’ and being ‘family’ is explicit. ‘Close’ (ngamu) refers to geographical or spatial proximity, contrasting with ‘distant’ (tiwa, warnma) or ‘far away’.
Sackett (1976, p. 142) documented that distance is not simply a criterion of affinal alliance or other kin relationship, but is integral to Western Desert kinship terminology—that is, it is encoded in kin terminology itself:
In addition to forming a closed system of relationship and behaviour, kinship determined marriage. It should be recalled that among the Kariera a man married a woman who he called by the term applied to an actual MBD (Radcliffe-Brown 1930, p. 48). In the Western Desert this rule was not applicable, for actual cross-cousins were kept terminologically distinct from more distant MBD/FZD (njuba) and called djudju/malyanj or yinganji (Tonkinson 1966, pp. 111–12). Likewise, male cross-cousins—the brothers of yinganji—were termed differently from their more distant counterparts. Occasionally they were addressed as gudja/maljanj, the same as siblings and parallel cousins, but most usually as wadjira or djamidi, meaning close cross-cousin of the same sex. The brothers of njuba—distant MBD/FZD—were called yungguri, maridji, or magunjdja.
It can hardly be argued that physical distance is not a primary consideration, and, in my view, the primary consideration upon which the structure of the Western Desert kinship system is built. Nevertheless, the issue that inevitably occurs in kinship description emerges—namely, the relationship between genealogical distance and physical distance, or, as has already been broached, whether the distinction between ‘close’ and ‘distant’ is a corollary of consanguineal and classificatory genealogical categorisation. The implication that arises from the latter possibility is that if distant kin are classificatory kin, that is, are only kin by virtue of adherence to an all-embracing and coherent kinship system that has only a relatively small and limited biological component, then there is every probability that the philosophical underpinning of the system is not genealogy but physical distance. In other words, genealogy as analogy becomes the means by which social relationships are enumerated. Most anthropologists (perhaps with the exception of Needham) have seen genealogical distance (for which, henceforth, the reader should interpret, unless otherwise specified, as biological genealogy) as integral to the interpretation of the close–distant dichotomy, even if this is seen as concurrent with physical distance. Dousset (2003, p. 53) provided the best description of the interrelationship between these two measures of distance in kinship in the Western Desert:
Sanctioned marriages among the Ngaatjatjarra are between cross cousins or between persons of the cross-cousin category two generations removed, such as classificatory MF or SS for female Ego, and a classificatory FM or DD for male Ego. Another jural marriage prescription is that marriage partners have to be ‘distant’, distance being measured in both genealogical and spatial terms. Genealogically speaking, a cousin has to be at least of the third degree to be a potential spouse. As genealogical memory does usually not exceed two generations, this means that the couple should not be able to trace a connecting genealogical link through their parents or grandparents. Spatially, wife and husband have distant geographical origins, with widely separated places of conception and birth, have not been prolonged co-residents prior to marriage, and are not associated with identical sites of significance. Hence, a bilateral cross-cousin prescription and a proximity proscription are operating. This is also reflected in people’s discourse, where geographic closeness is conflated with genealogical closeness, and where ‘coming from the same country’ is considered creating identical ‘consanguineal’ ties as being the common offspring of parents or grandparents. In terms of marriage descriptions and obligations as pronounced by indigenous people themselves, and in terms of modelling these descriptions, affiliation to land and genealogical structure cannot be disconnected. The genealogical aspect in the choice of spouses is described in terms of an obligation, hence a prescription, while the spatial part is pronounced as an interdiction, hence a proscription. Moreover, only about 2 per cent of marriages do not conform to these prescriptive and proscriptive rules.
It is clear from Dousset’s account that neither the genealogical nor spatial determinant acts separate to the other, that calculation of relationship depends on both. As Elkin’s verbatim report of the logic employed by his informant to arrive at an acceptable social compromise indicates, neither was an absolute, but perhaps existed more on a scale of social acceptability, with a large grey area of special circumstances and factors that could be tolerated if not too much was at stake:
Another woman who came to the camp was related to R__ as malan (younger sister). Her mother was own sister to his mother, and so was ngurndju (mother) to him. Moreover, as her husband was kamaru (mother’s brother) to R__, their marriage was from R__’s point of view a ‘brother-sister’ union. I was given to understand also that their marriage was a ‘little bit’ wrong because they both came from one ‘country’ (Oparina way), and had not observed the rule of local exogamy—an exceptional occurrence. If, however, R__’s kamaru had belonged to a different ‘country’ from that of his mother’s sister, all would have been in order, even though it would still have been a ‘brother-sister’ marriage. Of course such ‘brother’ and ‘sister’ may be distant cross-cousins, seeing that a mother’s brother’s female cross-cousin is his ‘sister’, Ego’s mother. (Elkin 1939–40, p. 218)
The question becomes one of sociocentric underpinning of the ideals that underlie Western Desert calculations as to the permissible, as distinct from the intolerable, and, deeper still, understanding why these principles exist in Western Desert social reckoning. Myers (1986, p. 175) dismissed one assumption for the Western Desert that is usually relevant to Aboriginal intergroup relationships: ‘Marriage constitutes one means of reproducing relatedness among individuals in a region’—that is, ‘the Pintupi prescription that spouses should come “from one country” cannot be reduced to the sociocentric formulation of marriage into a different band or descent group’.
Along with the extraordinary freedom that Western Desert people possessed in making personal arrangements outside their immediate circle (although, of course, these were not devoid of social and political considerations) came what appeared to be a consciousness of the Western Desert people and culture as a whole. In his consideration of what constituted the Australian ‘tribe’, Berndt (1959, p. 92) concluded of the Western Desert that ‘we might legitimately assume that there is a common awareness of belonging to a cultural and linguistic unit, over and above the smaller groups signified by these names, even though the actual span of the wider unit is not specified’. Berndt (1959, pp. 90–1, 103) described this Western Desert unity as a ‘social or cultural bloc’ with ‘no strict boundaries’ and within which ‘movements were relatively frequent’, with limitations on the breadth of the individual’s or local group’s involvement in the whole being imposed by distance: ‘People are accustomed to moving over a fairly large stretch of country, this was not by any means a matter of covering the whole cultural bloc’ (see Turner 1980, p. 9).
Sackett (1976, p. 142) noted that ‘Western Desert kinship formed a closed system, with ego related to all other persons in his social universe by actual or classificatory linkages’. He drew the conclusion that this was a ‘total system’, in which marriage based alliances were open-ended, such that ‘an alliance established between two groups by marriage could not be renewed or re-established for at least three generations’ (Sackett 1976, p. 146). Again, while physical restrictions meant the system could not have operated to produce social inculcation across its distribution, it was nonetheless a system predicated on the idea of a whole society, and not simply as an extension of the local group—a point that Myers (1986, p. 10) also made: ‘The Pintupi assert [that] they are all family’. As Myers (1986, p. 190) explained: ‘Each part, each local “unit”, can be produced only through cooperation of the larger structure. The organization of ceremony, requiring participation of others from far away, provides one way of constituting Pintupi society as a whole’. The role kinship plays in the organisation of this society is as ‘a structure that articulates the society not as a coordinated ordering of distinctive local groups, but as a set of related categories’.
The extent to which Western Desert people can be said to constitute ‘a society’ is debatable; most scholars of the Western Desert people would, I think, agree with Dousset (pers. comm., 24 April 2014): ‘I personally don’t think the entire Western Desert constitutes a society, but many societies’. This introduces another debate that will briefly be returned to in the conclusion, which is the role of kinship in the Western Desert people’s recent expansion into and across the Western Desert (see Dousset 2003; Hercus 1994, pp. 21–2; Holcombe 2004; McConvell 1996, 2001, p. 162; Myers 1986; Smith 2013, pp. 333–4; Strehlow 1947, pp. 61–2; Vincent 2011). In addition, there is a perception in anthropology, based on the vast expanse of its distribution and sparseness of its population density, and the divergent character of many of its social institutions, that Western Desert society represents a separate case among Aboriginal societies, or, at least, it is seen as an extremity in Aboriginal social organisation. Even in a world of harsh conditions, the Western Desert stands out as ranking among the absolute harshest. It is logical, therefore, to conclude that because Western Desert people were so thinly spread across a hostile environment, they had no option but to accentuate the physical distance calculus of their kin relationships. Nonetheless, social relationships designed not only to hold a society together and ensure its survival, but also to allow it to prosper in these conditions are characteristic of desert peoples generally (in respect to the Warlpiri, see Meggitt 1962, pp. 1, 49; Smith 2013, pp. 269–73). Western Desert kinship may represent an extreme manifestation of the close–distant dichotomy; however, it is one that features in Australian desert and arid societies more generally.
Turner’s (1974) Study of Groote Eylandt Kinship
At the other end of the distance scale, we find coastal societies for whom environmental conditions for the hunter-gatherer are at their most conducive, and, as a consequence, population densities at their highest. The evidence for the close–distant dichotomy in coastal peoples such as the Wik and Murinbata has already been examined, but two studies, in my view, have especial significance in the investigation of kinship distance in these societies. I refer to Turner’s (1974) Tradition and Transformation: A Study of Aborigines in the Groote Eylandt Area, Northern Australia and Rumsey’s (1981) ‘Kinship and context among the Ngarinyin’. Turner (1974, p. 16) made it apparent that distance is a crucial consideration in the reckoning of Wanungamagalyuagba kin relations:
It was found that knowledge of the terms used by an ego to refer to the two parents was not by itself enough for either I or an Aboriginal to work out what their children should be called. It was necessary to know, in addition, how each of the parents stood in relation to an ego in terms of whether they were his ‘close’ (augudangwa) or ‘far-away’ (auwilyagara) relatives. On discovering that ‘close’ and ‘far away’ were factors taken into account by ego in this decision process, informants were asked whether the relatives they had named in response were considered ‘close’ or ‘far-away’.
However, this was to prove a considerably different reckoning of distance than in, say, the Western Desert—both in its internal set of positive and negative conditioners, and because of the very different social structure to which it was applied. For the most part, Turner’s in-depth study concerns the four ‘hordes’ or local groups inhabiting Bickerton Island, a small island in the Gulf of Carpentaria between the mainland coast and the larger nearby Groote Eylandt (the study extends to include the larger orbit of Bickerton social life, taking in the mainland and Groote peoples). The Bickerton system has none of the freewheeling capacity to form marital alliances (either personally or structurally), as does the Western Desert system. Nonetheless, closer inspection reveals some similar kinship principles. While the Western Desert system excludes affinal alliance between kin known to share a common grandparent (which is pragmatically equated with common country), as does the Bickerton system. Like the Western Desert, consanguinity between the individual and their society was a matter of calculation: ‘“closeness” and “far-awayness” were reckoned in degrees’ (Turner 1974, p. 16):
Thus, members of Bickerton local groups consider each other ‘close’ relatives, even though one person may not actually have had a consanguineal relative in another’s local group within genealogical memory. The relative is considered ‘close’, however, because of an implied correspondence based on the belief that the Bickerton local groups intermarried in the distant past. Such a relative is considered less ‘close’ than one who has had an actual male or female ancestor located in one’s own local group. Here, the nearer this ancestor is to that ego’s own generation level, the ‘closer’ he considers the relative. On the other hand, a person belonging to a Wanindiliyuagwa or Nunggubuyu local group [i.e. of Groote Eylandt or the mainland respectively] is always considered ‘far-away’, unless he has a consanguine in an ego’s local group in living memory, or unless he is a local group mythically linked to that ego’s. (Turner 1974, p. 39)
The set of ideal relationships between the four Bickerton groups is mandated in the Nambirrirma myth: ‘The Wuramarba call the Wanungwadararbalangwa naningya [MMBDS] and dadingya [MMBDD]; Wuramara call the Wanungwadararbalangwa nabera [ZS] and dabera [ZD]’, and so on [all from the male propositus] (Turner 1974, p. 24 table 6). Thus, because ‘an ego’s “sons” are called nanugwa and a nanugwa’s “mother” is called dadingya; it is found that ego must call his own spouse dadingya. Knowing that Wuramarba calls Wanungwadararbalangwa dadingya, it can be concluded that Wurumarba’s spouse is from this local group’. In total then, ‘he [ego] calls the members of each local group by a different set of terms’ (Turner 1974, p. 26).5 Turner (1974, pp. 34–5) described the structural linkage between these groups as follows:
Aborigines see people as manifesting a substance and essence which is derived in part from their affiliation with a patrilineal local group through their father and in part from their affiliation with groups of their other consanguines. It is these interrelations that are, finally, the critical factors in their classification of kin … it is the fact that a particular individual belongs to a particular local group, say A, had a father (also A) born from a woman whose local group was B, and was himself born from a woman whose local group was C, who in turn was born from a woman whose group was D. This sociological-genealogical arrangement will be referred to as the ‘local group family’ and includes a person’s linkage to his own local group (through his father and father’s father), as well as to his father’s mother’s patrilineal local group, his mother’s, and his mother’s mother’s.
Genealogical reckoning for the Wanungamagalyuagba, unlike Western Desert society, is much more restricted and targeted, and can be seen as prescriptive rather than prohibitive. While Western Desert kinship does have its regional and residential subgroups, these are not determinative of relationship to anything like the degree found on Groote Eylandt and Bickerton Island. The Wanungamagalyuagba system is a relationship of the individual’s group, not the individual, to Bickerton Island’s society as a whole. Turner (1974, p. 3) reported: ‘It is the identification of men with their respective countries which seems to be at the basis of Wanungamagalyuagba social organisation, and indeed, the organization of all Aborigines in the Groote Eylandt area’. Personal identity is always given in terms of one’s country. Kin relationship is determined by a raft of factors that indicate proximity to the local group:
A ‘close’ relative is anyone who has or can be deduced to have had a consanguineal relative in an ego’s own local group. The nearer the relative is to an ego’s own generation level and the more similar his ‘local group family’ is to that of an ego’s, the ‘closer’ an ego will consider the ‘close’ relative in question. ‘Closeness’ is thus reckoned in terms of a certain type of correspondence between ‘local group families’. (Turner 1974, p. 38)
The Wanungamagalyuagba system is much more locally group-centric in its orientation than Western Desert groups at the other end of the scale—despite the fact that a central tenet of both systems is the prohibition of marriage within one’s group, axiomatic with one’s country. Desirable marriage for the Wanungamagalyuagba is seen as avoiding closeness, outright in respect to one’s own group, but also to any woman ‘whose forebears have recently formed some kind of marital alliance with the man’s local group’ (Turner 1974, pp. 39–44). The most desirable marriage partner will be that who is ‘really the most distant from him in terms of the relationship between her consanguines and people in his own local group’. However, ‘the most distant woman will still be one in a local group whose members exchanged women with, or took women from, an ego’s local group two generations ago’ (Turner 1974, p. 58). The reality is somewhat more ambiguous with the Bickerton Island groups, as is probably the case with coastal peoples generally (see Hiatt 1965, pp. 71–84):
Suppose the object were to prevent a man from marrying, first, someone in his own local group, second, someone in a group mythically linked to his, and third, someone outside these local groups whose consanguines were in his own local group, or in a local group linked to his. Under these circumstances and setting aside the intra-moiety prohibition for the moment it would be expected that ideally the woman he defined as ‘ideal wife’—always called dadingya—would be most distant in these terms … dadingya, however, would not be ‘farthest-away’ in strictly local group terms … Now relatives designated other than dadingya (e.g. denda, maminyamandja) may have consanguines in local groups only linked to an ego’s and be more desirable as wives from a ‘farthest-away’ point of view. Regardless of this … informants said that the ideal marriage would still be with a dadingya whose father’s mother was actually in one’s own local group as in the ‘only-four-local group’ situation. This indicates a definite preference for a woman in a local group into which one’s father’s father married and to which he gave his diyaba/dadiyamandja (‘sisters’) in return as wives. In other words, the maintenance of an exchange relationship between ego’s and another local group in alternate generations is preferred to acquiring a still ‘farther-away’ woman in local group terms. (Turner 1974, p. 92)
Unlike in the Western Desert, whereby acquisition of a ‘distant’ wife is good in unconditional terms, the object in the Wanungamagalyuagba case is the strengthening, or reinvigoration, of specified intergroup relations over time:
Marriage is then preferred with a relatively ‘close’ woman within the four [patrilineal local groups]. This ideological framework would seem to be a means of extending relationships outward to any number of alien groups yet of maintaining solidarity within a limited circle though encouraging ‘sister’-exchange between two groups in alternate generations, primarily on the level of the local group but occasionally on that of four more inclusive units (the complexes). This arrangement is formalised by the combination of two complexes into ‘somewhat “brother”-like groupings’ to form one exogamous moiety, but the links so-formed are not sufficiently strong to permit ‘sister’-exchange on a local group basis in consecutive generations. (Turner 1974, p. 98)
To summarise, the close–distant dichotomy among the Wanungamagalyuagba (and probably among coastal peoples generally) is not an open-ended desirability for ‘distant’ relationship, but a compromise between ‘closeness’ and ‘distance’: distant enough to be exogamous in respect of the local group and its recent attachments, but close enough to conform to a previous history of reciprocal interrelationship. This form of distance implies spatial distance—that is, the interrelationship of different groups and their countries; and genealogical distance, whereby certain genealogical relationships are prohibited on the grounds of being too recently enacted (e.g. an alliance between members of the same generation), while others, at the requisite temporal distance (two generations apart), are encouraged. As Turner (1974, p. 102) described it, what is required in ‘an ideal wife’ is one ‘who is relatively distant, sociologically’. For Turner, ‘this system may be seen as a compromise between the need for continually extending alliances over a wide range through obtaining wives from groups with no previous relationship to one’s own local group, and the need for stability and solidarity within a restricted circle of groups—in the interest of survival’.
Rumsey’s (1981) ‘Kinship and Context among the Ngarinyin’
Rumsey’s (1981) study of the kinship system of the Ngarinyin of the Kimberley region of Western Australia proceeded from a different perspective, with ostensibly different concerns; nevertheless, it has significant relevance for the close–distant dichotomy. In the Ngarinyin system, there is ‘the tendency for all persons within a single agnatic line to be called by the same kinterm’. In ego’s patriline, the usual Australian generational distinction is made: ‘Father (G + 1) and son (G –1) are called by the same term (idje), whereas father’s father (G + 2) and son’s son (G – 2) get called by the same terms as elder brother and younger brother (G + 0) respectively’ (p. 181). However, this is not the case in patrilines other than ego’s; for example, ‘if a man of one local clan or group be my “uncle”, kandingi (MB), then every man in it, irrespective of age, is my “uncle”: and every woman is classified as my “mother”, ngadji, being sister to kangingi’ (Elkin 1964, pp. 106–7 quoted in Rumsey 1981, p. 182). Therefore, we have what might be described as an instance of ‘extended skewing’, well beyond, for example, that of the mother’s brother/mother’s brother’s son conflation that is associated with Omaha skewing. Rumsey (1981, p. 182) qualified the application of this systematisation: ‘The Ngarinyin do not (at present anyway) think of or express relationships exclusively in those terms’. The close–distant dichotomy overlays this identification, operating in much the same terms as we have encountered in the studies already examined:
Marriage norms were expressed both positively and negatively: marrying within one’s moiety was traditionally punishable by death; marriages into the opposite moiety varied in degrees of correctness, depending on spouse’s kin class, and within each class, on socio-spatial-cum-genealogical ‘distance’ (distant relatives always being more highly valued for marriage than close ones). For a man these degrees ranged from ‘highly prescribed’ in the case of distant ‘father’s mother’ (a class which also included FMBD, FMBSD, etc.), down to ‘highly proscribed’ in the case of a close relative of the ‘mother’ class. (Rumsey 1981, p. 183)
Much the same rubric of relationship applies: notional (or classificatory) degrees of ‘consanguinity’ are tolerated, provided the person in question is at far enough spatial remove. Rumsey’s analysis of the Ngarinyin system draws these two threads together. Whereas we might think of the term ‘mother’ (ngaji) as the archetypically closest, and hence most proscribed, term of relationship, in its classificatory or distant aspect it implies the opposite: ‘The term ngaji, which in most contexts means “my (classificatory) mother”, was here being used as a cover term for “women of the opposite moiety”, mother being an especially salient exemplar of that class’ (Rumsey 1981, pp. 183–4). As such, it articulates a class for which a priori the investigation of affinal alliance is possible. Rumsey (1981, p. 184), investigating the context of these merged terminologies, uncovered further implications:
Maanggarra belongs to an interesting set of Ungarinyin terms, each of which refers to the set of clan estates associated with all of ego’s kinsmen of a given class. In the case of maanggarra, the relevant class is garndingi, and just when maanggarra is the topic of discourse, garndingi can be used to cover the entire range of kintypes which Elkin’s informants assigned to it, namely all the men of mother’s agnatic line, regardless of generation level. The reason for this is not difficult to discern: since clan membership is, in practice, determined by patrifiliation, all the members of any agnatic line—including those consisting of ego’s MF, MB, MBS, etc.—belong to the same clan. Hence in the discussion of maanggarra, the distinction normally implemented by the alternate-generation terms, garndingi and mamingi, becomes irrelevant, just as distinctions among alter-moiety female kin classes becomes irrelevant when the topic at hand was moiety exogamy.
At the outset of his discussion, Rumsey (1981, p. 181) referred to the ‘unusual’ status that anthropology had accorded the Ngarinyin kinship system in the past; Radcliffe-Brown (1930), Elkin (1931–32) and Scheffler (1978) are all noted as having regarded the Ngarinyin system as one that ‘differs significantly’ (Scheffler 1978, p. 417) from Australian norms. This, I believe, is not so, and hidden in the terminologies of eastern Australia collected by Fison, Howitt and others are many indications that a similar distinction is encoded between kinship reckoned among ‘close’ kin and separately in respect to ‘distant’ kin. This is a distinction that was observed as far back as Radcliffe-Brown (1930, p. 446): ‘Outside the circle of his immediate relatives he tends to classify other persons according to the hordes to which they belong. There are certain collective terms of relationship which the individual applies to different hordes. This tendency to treat their horde as a unit is … a determining factor of some importance in the Australian systems’. As Rumsey (1981, pp. 184–5) concluded in respect to the Ngarinyin system: ‘[There is] a high degree of correspondence between the generation-merging usage of the terms garndingi [MB] and ngaji [M], and contexts wherein the topic of discussion is interclan relations rather than intraclan or interpersonal ones … marriage … is in large measure conceived of as an interclan transaction’. Rumsey (1981, p. 185–6) only commented on the use of ‘vocative terms’ (e.g. ‘my father’ and ‘my mother’) in respect to close kin—terms that ‘are not nearly as apt to be used over “widened” ranges of kintypes as are the “referential” terms’. It is my belief that many as yet poorly studied Australian kinship systems have a distinction, both in terminology and structure based on the close–distant dichotomy, that is integral to them.
This brief and necessarily selective review of the close–distant dichotomy in the literature of Australian kinship has sought to address three key areas: 1) the pervasiveness of the dichotomy; 2) whether or not it is a genuine product of Indigenous thought; and 3) whether it is a matter of sentiment—of ideal—or whether distance is a determinative instrument in the structure (and, therefore, the terminology) of kinship.
The first point is not conclusive: although many of the prime authorities on Australian kinship have recognised close–distant dichotomy; equally, other authors make no mention of it at all. One factor in favour of the possibility that it represents a commonality in Australian kinship is its recorded presence in systems from various and diverse parts of the continent. Variance in application of the dichotomy in relation to distance and different standards of desirability, argues for an evolution that has gone hand in hand with the development of kin systems to fit historical and environmental circumstances.
The supposition that the close–distant dichotomy is an Indigenous conception is more certain. To my own knowledge, the dichotomy appears nowhere else in anthropology so frequently as it does in Australia. Nor can it be said to have a history of theoretical development within the discipline compared with descent theory or affinal theory. In most cases, it appears as though the anthropologist had been alerted to its importance by his or her informants, or observed it directly in action. There are, of course, numerous instances in which documentation of the dichotomy is noted directly from the informants, as quotation or case histories. In my view, there can be very little doubt that the close–distant dichotomy is a kinship principle through which Aboriginal people understand their own society and its interrelationships.
The last question—is the close-distant dichotomy structural (that is, not simply an ideal Aboriginal society’s cherish and strive towards, but actually encoded in structure)—is the least resolved of the three. Further work is necessary to demonstrate how the close–distant reckoning works in a kinship system, although many of the works cited in this chapter provide a good indication. I believe that enough has been shown in the sources reviewed to indicate the likelihood that kinship systems are founded on distance and that distance has a determinative role in the articulation of structure and terminology.
Some mention might be made of extensionist theory (otherwise, rewrite rule analysis, cf. Read 2001, pp. 243–4). Advocates of extensionist theory may argue that the close–distant dichotomy has already been well dealt with in the surmise that ‘fathers’ must inevitably extend from ‘a father’, the biological father, and so on, and that the dichotomy is structurally implied in Indigenous terminology to begin with. One only has to draw attention to Rumsey’s (1981) Ngarinyin example of ‘mothers’ who end up representing anything but the biological mother—in fact, one could say, the social antithesis of the biological mother (i.e. those who provide the key to those one is able to marry). Again, this question can hardly be given the breadth of consideration it deserves; however, it is significant that a champion of extensionist theory such as Shapiro (1979, p. 56) used the following example in discerning the difference between a ‘full’ father and ‘partial’ father:
Now consider that the adjectives dangang (‘full’) and marrkangga (‘partial’) can be used to modify any relationship term in this language—say, bapa (‘father’). It is ipso facto clear that a ‘full father’ is not only different from a ‘partial father’; he is more of a ‘father’ as well. And this is precisely the sort of subclassification that interests Scheffler. Who then is a ‘full’ father? When I first heard these adjectives used to modify relationship terms, I assumed that a ‘full’ member of any category is simply the occupant of that category who is genealogically closest to Ego. This assumption, I think, stemmed from a general ethnocentricism, as well as from rarer parochialism that pervades the culture of kinship buffs. Thus I was certain that a ‘full’ father is none other than one’s real, true, genuine and (above all) socially presumed father. But I was wrong. The Miwuyt [i.e. Yolngu] subcategory ‘full father’ does indeed include one’s genitor, but it embraces others as well—specifically, any ‘father’ who is a member of one’s genitor’s (and one’s own) ritual lodge. All other ‘fathers’ are ‘partial fathers’. Analogous notions apply to the subclassification of other Miwuyt categories.
In respect to Shapiro’s Yolngu example, at least, any tangible distinction between focal ‘fathers’ and extended ‘fathers’ (or any other relative) is based on the close–distant dichotomy that has been the subject of this chapter. Unlike extensionist theory, which is based purely on the application of logic, the close–distant dichotomy can be shown to be well founded in Indigenous thought and practical application. In my view, two related deductions follow: extensionist theory ends up becoming so all-embracing as to be effectively meaningless; and kinship structure has to do with the society, not the family, and biological designations as foci or anything else are irrelevant. In other words, analogy has become confused with aetiology. The implications of the close–distant dichotomy are indeed large and attempts to answer them will have to await a further forum.
Barnes, J 1965, ‘Foreword’, in L Hiatt (ed.), Kinship and conflict: a study of an Aboriginal community in northern Arnhem Land, Canberra: Australian National University Press, pp. vii–xi.
Berndt, RM 1959, ‘The concept of “the tribe” in the Western Desert of Australia’, Oceania, 30, pp. 88–106. doi.org/10.1002/j.1834-4461.1959.tb00213.x.
Berndt, RM & Berndt, CH 1970, Man, land and myth in North Australia: the Gunwinggu People, Sydney: Ure Smith.
Dousset, L 2003, ‘On the misinterpretation of the Aluridja Kinship system type (Australian Western Desert)’, Social Anthropology, 11(1), pp. 43–61. doi.org/10.1111/j.1469-8676.2003.tb00071.x.
Elkin, AP 1931–32, ‘The social organization of South Australian tribes’, Oceania, 2, pp. 44–73. doi.org/10.1002/j.1834-4461.1931.tb00022.x.
Elkin, AP 1937–38, ‘Kinship in South Australia’, Oceania, 8, pp. 419–25. doi.org/10.1002/j.1834-4461.1938.tb00434.x.
Elkin, AP 1938–39, ‘Kinship in South Australia’, Oceania, 9, pp. 41–78. doi.org/10.1002/j.1834-4461.1938.tb00216.x.
Elkin, AP 1939–40, ‘Kinship in South Australia’, Oceania, 10, pp. 196–234. doi.org/10.1002/j.1834-4461.1939.tb00276.x.
Falkenberg, A & Falkenberg, J 1981, The affinal relationship system, Oslo: Universitetsforlaget.
Hercus, L 1994, A grammar of the Arabana-Wangkangurru language, Lake Eyre Basin, South Australia, Pacific Linguistics Series C. No. 128, Canberra: Department of Linguistics, Research School of Pacific Studies, The Australian National University.
Hiatt, L 1965, Kinship and conflict: a study of an Aboriginal community in northern Arnhem Land, Canberra: Australian National University Press.
Holcombe, S 2004, ‘The politico-historical construction of the Pintupi Luritja and the concept of tribe’, Oceania, 74(4), pp. 257–75. doi.org/10.1002/j.1834-4461.2004.tb02854.x.
Howitt, AW 1904, The native tribes of south-east Australia, London: Macmillan & Co.
Kuper, A 2005, The reinvention of primitive society, London: Routledge.
Lévi-Strauss, C 1969, The elementary structures of kinship, Boston: Beacon Press.
McConnel, U 1933–34, ‘“The Wik-Munkan and allied tribes of Cape York Peninsula, N.Q.”. Part III—kinship and marriage’, Oceania, 4, pp. 310–67. doi.org/10.1002/j.1834-4461.1934.tb00113.x.
McConnel, U 1939–40, ‘Social organization of the tribes of Cape York Peninsula’, Oceania, 10, pp. 434–55. doi.org/10.1002/j.1834-4461.1940.tb00305.x.
McConvell, P 1996, ‘Backtracking to Babel: the chronology of Pama-Nyungan expansion in Australia’, Archaeology in Oceania, 31, pp. 125–44. doi.org/10.1002/j.1834-4453.1996.tb00356.x.
McConvell, P 2001, ‘Language shift and language spread among hunter-gatherers’, in C Panter-Brick, RH Layton & P Rowley-Conwy (eds), Hunter-gatherers: an interdisciplinary perspective, Cambridge: Cambridge University Press, pp. 144–67.
Meggitt, M 1962, Desert people, Chicago: University of Chicago Press.
Morgan, LH 1871, Systems of consanguinity and affinity in the human family, Washington: Smithsonian Institution.
Myers, FR 1986, Pintupi country, Pintupi self, Canberra: Australian Institute of Aboriginal Studies.
Needham, R 1962, ‘Genealogy and category in Wikmunkan society’, Ethnology, 1(2), pp. 223–64. doi.org/10.2307/3772877.
Radcliffe-Brown, AR 1913, ‘Three tribes of Western Australia’, Journal of the Royal Anthropological Institute, 43, pp. 143–94.
Radcliffe-Brown, AR 1930, ‘The social organization of Australian tribes: Part I’, Oceania, 1(1), pp. 34–63. doi.org/10.1002/j.1834-4461.1930.tb00003.x.
Read, DW 2001, ‘Formal analysis of kinship terminologies and its relationship to what constitutes kinship’, Anthropological Theory, 1(2), pp. 239–67. doi.org/10.1177/146349960100100205.
Rumsey, A 1981, ‘Kinship and context among the Ngarinyin’, Oceania, 51, pp. 181–92. doi.org/10.1002/j.1834-4461.1981.tb01448.x.
Sackett, L 1976, ‘Indirect exchange in a symmetrical system: marriage alliance in the Western Desert of Australia’, Ethnology, 15(2), pp. 135–49. doi.org/10.2307/3773325.
Scheffler, HW 1978, Australian kin classification, Cambridge: Cambridge University Press. doi.org/10.1017/CBO9780511557590.
Shapiro, W 1979, Social organization in Aboriginal Australia, Canberra: Australian National University Press.
Smith, M 2013, The archaeology of Australia’s deserts, Cambridge: Cambridge University Press. doi.org/10.1017/CBO9781139023016.
Stanner, WEH 1935–36, ‘A note on Djamindjung kinship and totemism’, Oceania, 6, pp. 441–51. doi.org/10.1002/j.1834-4461.1936.tb00204.x.
Stanner, WEH 1936–37, ‘Murinbata kinship and totemism’, Oceania, 7, pp. 186–216. doi.org/10.1002/j.1834-4461.1936.tb00451.x.
Strehlow, TGH 1947, Aranda traditions, Melbourne University Press [Reprinted 1968, New York: Johnson Reprint Corporation].
Sutton, P 1978, Wik: Aboriginal society, territory and language at Cape Keerweer, Cape York Peninsula, Australia, PhD thesis, University of Queensland.
Turner, DH 1974, Tradition and transformation: a study of Aborigines in the Groote Eylandt area, northern Australia, Australian Aboriginal Studies No. 53, Canberra: Australian Institute of Aboriginal Studies.
Turner, DH 1980, Australian Aboriginal social organization, Atlantic Highlands, NJ: Humanities Press International.
Vincent, E 2011, ‘The everyday effects of native title: attending to Indigenous self-accounts’, Australian Anthropological Society conference paper, Brisbane.
1 An exception is Radcliffe-Brown (1930, pp. 438–9), of whom Turner (1980, p. ix–x) commented: ‘The theoretical implications of the concepts “close” and “far-away” were anticipated but never fully explored by Radcliffe-Brown’.
2 There is much that could be teased out of Elkin’s work on the distance dichotomy; however, space does not permit such an excursion here. Elkin’s interpretation of the motivation for distant affinal relationships in this passage—that Male Ego seeks a distant mother-in-law for the sake of obviating rigorous avoidance strictures with close kin—is, in my view, an insufficient explanation for the dichotomy.
3 This distinction hardly applies to anthropologists without exception. For example, Hiatt (1965) made use of expressions such as ‘close classificatory’ and ‘distant classificatory’ relatives throughout his work, without making clear on what basis the distinction was made (for examples of the former, see Hiatt 1965, pp. 96–8). Barnes (1965, p. viii) alluded to Hiatt’s failure to define this distinction (which, notwithstanding, recurs frequently throughout his work) in his foreword: ‘Most descriptions of Aboriginal marriage arrangements, and of involvement in quarrels and fights, are in terms of ties of “close” or “distant” kinship of one kind or another. The reader is usually left in the dark about the exact meaning of “close” and “distant”, if one exists’.
4 Stanner (1935–36, pp. 443–4) made a similar, and perhaps more telling, distinction in respect to Djamindjung kinship: ‘Classificatory mother’s brother’s daughters and father’s sister’s daughters may be married, and marriages seem to be allowed with distant mother’s mother’s brother’s son’s daughters, distant classificatory sisters, and distant classificatory mothers’.
5 This is very much like the Murinbata system, as described by Falkenberg and Falkenberg (1981, pp. 143–5), in which, similarly, the four ‘patrilineal descent lines’ that comprise Murinbata society are each composed of different, generationally skewed, terms that distinguish them from Ego’s own ‘close’ ‘patrilineal’ group. Hiatt’s (1965, pp. 44–46, 50) tentative (to use his term ‘hypothetical’) reconstruction of Gidjingali kinship in terms of the interrelationship between four patrilineal groups, and his description of the interaction of ‘communities’ points to much of the same sort of relationship (see Hiatt 1965, pp. 25–6, 33).